Closed SverreNystad closed 1 year ago
OpenAIs docs on image generation: https://platform.openai.com/docs/guides/images
An solution is to use DALL-E to generate pictures and have the GPT just make the prompt for DALL-E given the description
First image generated
generate_image(prompt="A painting of a dungeon master sitting at a table with a group of adventurers playing Dungeons and Dragons.") ->
The generation of the image did not take very long time, This can make this part of the system easier to use in runtime
While the GPT Dungeon Master provides rich and immersive textual descriptions, visualizing these descriptions can enhance the player's experience and immersion.
I propose the integration of a Generative Art feature that can translate the GPT DM's textual descriptions into real-time visual art. This would allow players to not only hear or read about a scene, character, or item but also see an artistic representation generated on-the-fly.
Proposed Features: